On Wed, Apr 16, 2008 at 12:15:08PM -0700, Christoph Lameter wrote:
> On Wed, 16 Apr 2008, Robin Holt wrote:
>
> > On Wed, Apr 16, 2008 at 11:35:38AM -0700, Christoph Lameter wrote:
> > > On Wed, 16 Apr 2008, Robin Holt wrote:
> > >
> > > > I don't
On Thu, Apr 17, 2008 at 05:51:57PM +0200, Andrea Arcangeli wrote:
> On Wed, Apr 16, 2008 at 11:35:38AM -0700, Christoph Lameter wrote:
> > On Wed, 16 Apr 2008, Robin Holt wrote:
> >
> > > I don't think this lock mechanism is completely working. I have
> >
On Thu, Apr 17, 2008 at 07:14:43PM +0200, Andrea Arcangeli wrote:
> On Thu, Apr 17, 2008 at 11:36:42AM -0500, Robin Holt wrote:
> > In this case, we are not making the call to unregister, we are waiting
> > for the _release callout which has already removed it from the list.
> &
On Tue, Apr 22, 2008 at 02:00:56PM +0200, Andrea Arcangeli wrote:
> On Tue, Apr 22, 2008 at 09:20:26AM +0200, Andrea Arcangeli wrote:
> > invalidate_range_start {
> > spin_lock(&kvm->mmu_lock);
> >
> > kvm->invalidate_range_count++;
> > rmap-invalidate of sptes in range
> >
>
>
On Tue, Apr 22, 2008 at 03:21:43PM +0200, Andrea Arcangeli wrote:
> On Tue, Apr 22, 2008 at 08:01:20AM -0500, Robin Holt wrote:
> > On Tue, Apr 22, 2008 at 02:00:56PM +0200, Andrea Arcangeli wrote:
> > > On Tue, Apr 22, 2008 at 09:20:26AM +0200, Andr
n_one, and try_to_unmap_one call sites.
On Tue, Apr 22, 2008 at 03:48:47PM +0200, Andrea Arcangeli wrote:
> On Tue, Apr 22, 2008 at 08:36:04AM -0500, Robin Holt wrote:
> > I am a little confused about the value of the seq_lock versus a simple
> > atomic, but I assumed there is a r
leeping version less acceptable.
Can we agree upon this list of issues?
Thank you,
Robin Holt
-
This SF.net email is sponsored by the 2008 JavaOne(SM) Conference
Don't miss this year's exciting event. There's sti
On Tue, Apr 22, 2008 at 08:43:35PM +0200, Andrea Arcangeli wrote:
> On Tue, Apr 22, 2008 at 01:22:13PM -0500, Robin Holt wrote:
> > 1) invalidate_page: You retain an invalidate_page() callout. I believe
> > we have progressed that discussion to the point that it requires some
&g
On Tue, Apr 22, 2008 at 01:19:29PM -0700, Christoph Lameter wrote:
> Thanks for adding most of my enhancements. But
>
> 1. There is no real need for invalidate_page(). Can be done with
> invalidate_start/end. Needlessly complicates the API. One
> of the objections by Andrew was that t
> The only other change I did has been to move mmu_notifier_unregister
> at the end of the patchset after getting more questions about its
> reliability and I documented a bit the rmmod requirements for
> ->release. we'll think later if it makes sense to add it, nobody's
> using it anyway.
XPMEM i
On Wed, Apr 23, 2008 at 12:43:52AM +0200, Andrea Arcangeli wrote:
> On Tue, Apr 22, 2008 at 01:22:55PM -0700, Christoph Lameter wrote:
> > Looks like this is not complete. There are numerous .h files missing which
> > means that various structs are undefined (fs.h and rmap.h are needed
> > f.e.)
On Wed, Apr 23, 2008 at 03:36:19PM +0200, Andrea Arcangeli wrote:
> On Tue, Apr 22, 2008 at 06:07:27PM -0500, Robin Holt wrote:
> > > The only other change I did has been to move mmu_notifier_unregister
> > > at the end of the patchset after getting more questions about its
&g
On Wed, Apr 23, 2008 at 03:44:27PM +0200, Andrea Arcangeli wrote:
> On Tue, Apr 22, 2008 at 04:14:26PM -0700, Christoph Lameter wrote:
> > We want a full solution and this kind of patching makes the patches
> > difficuilt to review because later patches revert earlier ones.
>
> I know you rather
On Wed, Apr 23, 2008 at 06:15:45PM +0200, Andrea Arcangeli wrote:
> Once I get confirmation that everyone is ok with #v13 I'll push a #v14
> before Saturday with that cosmetical error cleaned up and
> mmu_notifier_unregister moved at the end (XPMEM will have unregister
> don't worry). I expect the
I am not certain of this, but it seems like this patch leaves things in
a somewhat asymetric state. At the very least, I think that asymetry
should be documented in the comments of either mmu_notifier.h or .c.
Before I do the first mmu_notifier_register, all places that test for
mm_has_notifiers(
On Fri, Apr 25, 2008 at 06:56:40PM +0200, Andrea Arcangeli wrote:
> Fortunately I figured out we don't really need mm_lock in unregister
> because it's ok to unregister in the middle of the range_begin/end
> critical section (that's definitely not ok for register that's why
> register needs mm_lock
On Thu, Apr 24, 2008 at 07:41:45PM +0200, Andrea Arcangeli wrote:
> A full new update will some become visible here:
>
>
> http://www.kernel.org/pub/linux/kernel/people/andrea/patches/v2.6/2.6.25/mmu-notifier-v14-pre3/
I grabbed these and built them. Only change needed was another include
> I however doubt this will bring us back to the same performance of the
> current spinlock version, as the real overhead should come out of
> overscheduling in down_write ai anon_vma_link. Here an initially
> spinning lock would help but that's gray area, it greatly depends on
> timings, and on ve
> diff --git a/mm/Kconfig b/mm/Kconfig
> --- a/mm/Kconfig
> +++ b/mm/Kconfig
> @@ -205,3 +205,6 @@ config VIRT_TO_BUS
> config VIRT_TO_BUS
> def_bool y
> depends on !ARCH_NO_VIRT_TO_BUS
> +
> +config MMU_NOTIFIER
> + bool
Without some text following the bool keyword, I am not even
On Mon, May 05, 2008 at 12:08:25AM +0200, Andrea Arcangeli wrote:
> On Sun, May 04, 2008 at 02:13:45PM -0500, Robin Holt wrote:
> > > diff --git a/mm/Kconfig b/mm/Kconfig
> > > --- a/mm/Kconfig
> > > +++ b/mm/Kconfig
> > > @@ -205,3 +205,6 @@ confi
You can drop this patch.
This turned out to be a race in xpmem. It "appeared" as if it were a
race in get_task_mm, but it really is not. The current->mm field is
cleared under the task_lock and the task_lock is grabbed by get_task_mm.
I have been testing you v15 version without this patch and n
ide note, we currently have XPMEM working on x86_64 SSI, and
ia64 cross-partition. We are in the process of getting XPMEM working
on x86_64 cross-partition in support of UV.
Thanks,
Robin Holt
-
This SF.net email is sponsored by
ust using stop-machine is actually the right one just
> because it's _so_ simple.
Also, stop-machine will not work if we come to the conclusion that
i_mmap_lock and anon_vma->lock need to be sleepable locks.
Thanks,
Robin Holt
---
On Tue, May 13, 2008 at 10:06:44PM +1000, Nick Piggin wrote:
> On Thursday 08 May 2008 10:38, Robin Holt wrote:
> > In order to invalidate the remote page table entries, we need to message
> > (uses XPC) to the remote side. The remote side needs to acquire the
> > importing p
On Wed, May 14, 2008 at 06:11:22AM +0200, Nick Piggin wrote:
> On Tue, May 13, 2008 at 10:32:38AM -0500, Robin Holt wrote:
> > On Tue, May 13, 2008 at 10:06:44PM +1000, Nick Piggin wrote:
> > > On Thursday 08 May 2008 10:38, Robin Holt wrote:
> > > > In order to in
On Wed, May 14, 2008 at 08:18:21AM -0700, Linus Torvalds wrote:
>
>
> On Wed, 14 May 2008, Robin Holt wrote:
> >
> > Are you suggesting the sending side would not need to sleep or the
> > receiving side?
>
> One thing to realize is that most of the time (read: pre
We are pursuing Linus' suggestion currently. This discussion is
completely unrelated to that work.
On Thu, May 15, 2008 at 09:57:47AM +0200, Nick Piggin wrote:
> I'm not sure if you're thinking about what I'm thinking of. With the
> scheme I'm imagining, all you will need is some way to raise an
On Fri, May 16, 2008 at 01:52:03AM +0200, Nick Piggin wrote:
> On Thu, May 15, 2008 at 10:33:57AM -0700, Christoph Lameter wrote:
> > On Thu, 15 May 2008, Nick Piggin wrote:
> >
> > > Oh, I get that confused because of the mixed up naming conventions
> > > there: unmap_page_range should actually b
On Fri, May 16, 2008 at 06:23:06AM -0500, Robin Holt wrote:
> On Fri, May 16, 2008 at 01:52:03AM +0200, Nick Piggin wrote:
> > On Thu, May 15, 2008 at 10:33:57AM -0700, Christoph Lameter wrote:
> > > On Thu, 15 May 2008, Nick Piggin wrote:
> > >
> > > >
e cleanup is equivalent to the cleanup of
a page we have never used.
Thanks,
Robin Holt
-
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.at
On Wed, Jan 23, 2008 at 12:27:57PM +0200, Avi Kivity wrote:
>> The approach with the export notifier is page based not based on the
>> mm_struct. We only need a single page count for a page that is exported to
>> a number of remote instances of linux. The page count is dropped when all
>> the remot
Christoph, Maybe you can clear one thing up. Was this proposal an
addition to or replacement of Andrea's? I assumed an addition. I am
going to try to restrict my responses to ones appropriate for that
assumption.
> The remote instance is like a secondary TLB what you're doing in your
> code is
> Why don't you stick to mm+va and use get_user_pages and let the VM do
> the swapins etc...?
Our concern is also with memory protections for the physical page.
Additionally, we need to support exporting of memory in a VM_PFNMAP
address ranges (specifically, memory mapped using sgi_fetchop, uncach
On Wed, Jan 23, 2008 at 01:51:23PM +0100, Gerd Hoffmann wrote:
> Jumping in here, looks like this could develop into a direction useful
> for Xen.
>
> Background: Xen has a mechanism called "grant tables" for page sharing.
> Guest #1 can issue a "grant" for another guest #2, which in turn then
>
On Wed, Jan 23, 2008 at 03:12:36PM +0100, Gerd Hoffmann wrote:
> Hi,
>
> >> That would render the notifies useless for Xen too. Xen needs to
> >> intercept the actual pte clear and instead of just zapping it use the
> >> hypercall to do the unmap and release the grant.
> >
> > We are tackling
On Wed, Jan 23, 2008 at 03:35:54PM +0100, Gerd Hoffmann wrote:
> Robin Holt wrote:
> > We have a seg structure which is similar to some structure you probably
> > have which describes the grant. One of the things hanging off that
> > seg structure is essentially a page table
On Wed, Jan 23, 2008 at 11:48:43AM -0800, Christoph Lameter wrote:
> On Wed, 23 Jan 2008, Andrea Arcangeli wrote:
>
> > On Wed, Jan 23, 2008 at 04:52:47AM -0600, Robin Holt wrote:
> > > But 100 callouts holding spinlocks will not work for our implementation
> > > and
early Friday. This has not even been compiled yet. Just
marking it up for now.
Thank you for your attention,
Robin Holt
Index: mmu_notifiers/include/linux/export_notifier.h
===
--- /dev/null 1970-01-01 00:00:00.0 +
getting
to the point where we can start testing. It does compile now.
I am traveling tomorrow but should be able to get back to this tomorrow
evening or early Friday.
Thank you for your attention,
Robin Holt
Index: mmu_notifiers/include/linux/export_notifier.h
On Fri, Jan 25, 2008 at 11:03:07AM -0800, Christoph Lameter wrote:
> > > > Shouldn't this really be protected by the down_write(mmap_sem)? Maybe:
> > >
> > > Ok. We could switch this to mmap_sem protection for the mm_struct but the
> > > rmap notifier is not associated with an mm_struct. So we w
On Fri, Jan 25, 2008 at 12:42:29PM +0100, Andrea Arcangeli wrote:
> On a technical merit this still partially makes me sick and I think
> it's the last issue to debate.
>
> @@ -971,6 +974,9 @@ int try_to_unmap(struct page *page, int
> else
> ret = try_to_unmap_file(page, m
> +#define mmu_notifier(function, mm, args...) \
...
> + if (__mn->ops->function)\
> + __mn->ops->function(__mn, \
> + mm, \
On Fri, Jan 25, 2008 at 10:47:04AM -0800, Christoph Lameter wrote:
> On Fri, 25 Jan 2008, Robin Holt wrote:
>
> > I realize it is a minor nit, but since we put the continuation in column
> > 81 in the next define, can we do the same here and make this more
> > readable?
> void mmu_notifier_register(struct mmu_notifier *mn, struct mm_struct *mm)
> {
> - spin_lock(&mmu_notifier_list_lock);
> - hlist_add_head(&mn->hlist, &mm->mmu_notifier.head);
> - spin_unlock(&mmu_notifier_list_lock);
> + down_write(&mm->mmap_sem);
> + __mmu_notifier_register(
> > > 1. invalidate_all()
> >
> > That will be fine as long as we can unregister the ops notifier and free
> > the structure. Otherwise, we end up being called needlessly.
>
> No you cannot do that because there are still callbacks that come later.
> The invalidate_all may lead to invalidate_ra
> +void mmu_notifier_release(struct mm_struct *mm)
...
> + hlist_for_each_entry_safe_rcu(mn, n, t,
> + &mm->mmu_notifier.head, hlist) {
> + if (mn->ops->release)
> + mn->ops->release(mn, mm);
> +
I am going to seperate my comments into individual replies to help
reduce the chance they are lost.
> +void mmu_notifier_release(struct mm_struct *mm)
...
> + hlist_for_each_entry_safe_rcu(mn, n, t,
> + &mm->mmu_notifier.head, hlist) {
> +
What is the status of getting invalidate_all adjusted to indicate a need
to also call _release?
Thanks,
Robin
On Mon, Jan 28, 2008 at 12:28:46PM -0800, Christoph Lameter wrote:
> when a task exits we can remove all external pts at once. At that point the
> extern mmu may also unregister itself fr
I don't understand how this is intended to work. I think the page flag
needs to be maintained by the mmu_notifier subsystem.
Let's assume we have a mapping that has a grant from xpmem and an
additional grant from kvm. The exporters are not important, the fact
that there may be two is.
Assume th
> Robin, if you don't mind, could you please post or upload somewhere
> your GPLv2 code that registers itself in Christoph's V2 notifiers? Or
> is it top secret? I wouldn't mind to have a look so I can better
> understand what's the exact reason you're sleeping besides attempting
> GFP_KERNEL alloc
On Wed, Jan 30, 2008 at 06:04:52PM +0100, Andrea Arcangeli wrote:
> On Wed, Jan 30, 2008 at 10:11:24AM -0600, Robin Holt wrote:
...
> > The three issues we need to simultaneously solve is revoking the remote
> > page table/tlb information while still in a sleepable context and not
Back to one of Andrea's points from a couple days ago, I think we still
have a problem with the PageExternalRmap page flag.
If I had two drivers with external rmap implementations, there is no way
I can think of for a simple flag to coordinate a single page being
exported and maintained by the two
This is the second part of a patch posted to patch 1/6.
Index: git-linus/mm/rmap.c
===
--- git-linus.orig/mm/rmap.c2008-01-30 11:55:56.0 -0600
+++ git-linus/mm/rmap.c 2008-01-30 12:01:28.0 -0600
@@ -476,8 +476,10
On Wed, Jan 30, 2008 at 11:50:26AM -0800, Christoph Lameter wrote:
> On Wed, 30 Jan 2008, Andrea Arcangeli wrote:
>
> > XPMEM requires with invalidate_range (sleepy) +
> > before_invalidate_range (sleepy). invalidate_all should also be called
> > before_release (both sleepy).
> >
> > It sounds we
On Wed, Jan 30, 2008 at 11:19:28AM -0800, Christoph Lameter wrote:
> On Wed, 30 Jan 2008, Jack Steiner wrote:
>
> > Moving to a different lock solves the problem.
>
> Well it gets us back to the issue why we removed the lock. As Robin said
> before: If its global then we can have a huge number o
> Well the GRU uses follow_page() instead of get_user_pages. Performance is
> a major issue for the GRU.
Worse, the GRU takes its TLB faults from within an interrupt so we
use follow_page to prevent going to sleep. That said, I think we
could probably use follow_page() with FOLL_GET set to acco
On Thu, Jan 31, 2008 at 05:37:21PM -0800, Christoph Lameter wrote:
> On Fri, 1 Feb 2008, Andrea Arcangeli wrote:
>
> > I appreciate the review! I hope my entirely bug free and
> > strightforward #v5 will strongly increase the probability of getting
> > this in sooner than later. If something else
On Thu, Jan 31, 2008 at 07:56:12PM -0600, Jack Steiner wrote:
> > @@ -2033,6 +2034,7 @@ void exit_mmap(struct mm_struct *mm)
> > unsigned long end;
> >
> > /* mm's last user has gone, and its about to be pulled down */
> > + mmu_notifier(invalidate_all, mm, 0);
> > arch_exit_mmap(mm
Christoph,
Jack has repeatedly pointed out needing an unregister outside the
mmap_sem. I still don't see the benefit to not having the lock in the mm.
Thanks,
Robin
Index: mmu_notifiers-cl-v4/include/linux/mm_types.h
===
--- mmu_no
On Thu, Jan 31, 2008 at 05:57:25PM -0800, Christoph Lameter wrote:
> Move page tables also needs to invalidate the external references
> and hold new references off while moving page table entries.
I must admit to not having spent any time thinking about this, but aren't
we moving the entries from
On Thu, Jan 31, 2008 at 06:39:19PM -0800, Christoph Lameter wrote:
> On Thu, 31 Jan 2008, Robin Holt wrote:
>
> > Jack has repeatedly pointed out needing an unregister outside the
> > mmap_sem. I still don't see the benefit to not having the lock in the mm.
>
> I ne
> Index: linux-2.6/include/linux/mmu_notifier.h
...
> +struct mmu_notifier_ops {
...
> + /*
> + * invalidate_range_begin() and invalidate_range_end() must paired.
> + * Multiple invalidate_range_begin/ends may be nested or called
> + * concurrently. That is legit. However, no new
On Thu, Jan 31, 2008 at 07:58:40PM -0800, Christoph Lameter wrote:
> On Thu, 31 Jan 2008, Robin Holt wrote:
>
> > > + void (*invalidate_range_end)(struct mmu_notifier *mn,
> > > + struct mm_struct *mm, int atomic);
> >
> > I think w
> Index: linux-2.6/mm/memory.c
...
> @@ -1668,6 +1678,7 @@ gotten:
> page_cache_release(old_page);
> unlock:
> pte_unmap_unlock(page_table, ptl);
> + mmu_notifier(invalidate_range_end, mm, 0);
I think we can get an _end call without the _begin call before it.
Thanks,
Robi
do_wp_page can reach the _end callout without passing the _begin
callout. This prevents making the _end unles the _begin has also
been made.
Index: mmu_notifiers-cl-v5/mm/memory.c
===
--- mmu_notifiers-cl-v5.orig/mm/memory.c2
On Fri, Feb 01, 2008 at 04:32:21AM -0600, Robin Holt wrote:
> On Thu, Jan 31, 2008 at 08:43:58PM -0800, Christoph Lameter wrote:
> > On Thu, 31 Jan 2008, Robin Holt wrote:
> >
> > > > Index: linux-2.6/mm/memory.c
> > > ...
On Thu, Jan 31, 2008 at 08:43:58PM -0800, Christoph Lameter wrote:
> On Thu, 31 Jan 2008, Robin Holt wrote:
>
> > > Index: linux-2.6/mm/memory.c
> > ...
> > > @@ -1668,6 +1678,7 @@ gotten:
> > > page_cache_release(old_page);
> > >
OK. Now that release has been moved, I think I agree with you that the
down_write(mmap_sem) can be used as our lock again and still work for
Jack. I would like a ruling from Jack as well.
Thanks,
Robin
-
This SF.net email i
On Fri, Feb 01, 2008 at 04:55:16AM -0600, Robin Holt wrote:
> OK. Now that release has been moved, I think I agree with you that the
> down_write(mmap_sem) can be used as our lock again and still work for
> Jack. I would like a ruling from Jack as well.
Ignore this, I was in the wrong
With this set of patches, I think we have enough to get xpmem working
with most types of mappings. In the past, we operated without any of
these callouts by significantly restricting why types of mappings could
remotely fault and what types of operations the user could do. With
this set, I am ce
Argh,
Here is the correct one. Sorry
On Fri, Feb 01, 2008 at 05:58:41AM -0600, Robin Holt wrote:
> With this set of patches, I think we have enough to get xpmem working
> with most types of mappings. In the past, we operated without any of
> these callouts by significantly restri
Christoph,
The following code in do_wp_page is a problem.
We are getting this callout when we transition the pte from a read-only
to read-write. Jack and I can not see a reason we would need that
callout. It is causing problems for xpmem in that a write fault goes
to get_user_pages which gets b
On Fri, Feb 01, 2008 at 03:19:32PM -0800, Christoph Lameter wrote:
> On Fri, 1 Feb 2008, Robin Holt wrote:
>
> > We are getting this callout when we transition the pte from a read-only
> > to read-write. Jack and I can not see a reason we would need that
> > callout. It
On Fri, Feb 01, 2008 at 04:05:08PM -0800, Christoph Lameter wrote:
> On Fri, 1 Feb 2008, Robin Holt wrote:
>
> > On Fri, Feb 01, 2008 at 03:19:32PM -0800, Christoph Lameter wrote:
> > > On Fri, 1 Feb 2008, Robin Holt wrote:
> > >
> > > > We are getti
On Fri, Feb 01, 2008 at 06:21:45PM -0600, Robin Holt wrote:
> On Fri, Feb 01, 2008 at 04:05:08PM -0800, Christoph Lameter wrote:
> > Are you saying that you get the callback when transitioning from a read
> > only to a read write pte on the *same* page?
>
> I believe tha
Great news! I have taken over Dean's xpmem patch set while he is on
sabbatical. Before he left, he had his patch mostly working on top of
this patch set. We had one deadlock. I have coded for that specific
deadlock and xpmem now passes a simple grant/attach/fault/fork/unmap/map
test.
After ana
On Tue, Feb 05, 2008 at 02:06:23PM -0800, Christoph Lameter wrote:
> On Tue, 5 Feb 2008, Andrea Arcangeli wrote:
>
> > On Tue, Feb 05, 2008 at 10:17:41AM -0800, Christoph Lameter wrote:
> > > The other approach will not have any remote ptes at that point. Why would
> > > there be a coherency issu
On Fri, Feb 08, 2008 at 03:32:19PM -0800, Christoph Lameter wrote:
> On Fri, 8 Feb 2008, Andrew Morton wrote:
>
> > What about ib_umem_get()?
>
> Ok. It pins using an elevated refcount. Same as XPmem right now. With that
> we effectively pin a page (page migration will fail) but we will
> conti
On Fri, Feb 08, 2008 at 03:41:24PM -0800, Christoph Lameter wrote:
> On Fri, 8 Feb 2008, Robin Holt wrote:
>
> > > > What about ib_umem_get()?
>
> Correct.
>
> You missed the turn of the conversation to how ib_umem_get() works.
> Currently it seems to pin t
On Thu, Feb 14, 2008 at 09:09:08AM -0600, Steve Wise wrote:
> Note that for T3, this involves suspending _all_ rdma connections that are
> in the same PD as the MR being remapped. This is because the driver
> doesn't know who the application advertised the rkey/stag to. So without
Is there a
On Sat, Feb 16, 2008 at 11:48:27AM +0100, Andrea Arcangeli wrote:
> Those below two patches enable KVM to swap the guest physical memory
> through Christoph's V7.
>
> There's one last _purely_theoretical_ race condition I figured out and
> that I'm wondering how to best fix. The race condition wor
On Sun, Feb 17, 2008 at 04:01:20AM +0100, Andrea Arcangeli wrote:
> On Sat, Feb 16, 2008 at 11:21:07AM -0800, Christoph Lameter wrote:
> > On Fri, 15 Feb 2008, Andrew Morton wrote:
> >
> > > What is the status of getting infiniband to use this facility?
> >
> > Well we are talking about this it s
On Tue, Feb 19, 2008 at 09:43:57AM +0100, Nick Piggin wrote:
> So I implemented mmu notifiers slightly differently. Andrea's mmu notifiers
> are rather similar. However I have tried to make a point of minimising the
> impact the the core mm/. I don't see why we need to invalidate or flush
> anythin
On Wed, Feb 20, 2008 at 01:52:06AM +0100, Andrea Arcangeli wrote:
> On Wed, Feb 20, 2008 at 12:04:27AM +0100, Nick Piggin wrote:
> > OK (thanks to Robin as well). Now I understand why you are using it,
> > but I don't understand why you don't defer new TLBs after the point
> > where the linux pte c
On Wed, Feb 20, 2008 at 12:11:57AM +0100, Nick Piggin wrote:
> On Tue, Feb 19, 2008 at 02:58:51PM +0100, Andrea Arcangeli wrote:
> > On Tue, Feb 19, 2008 at 09:43:57AM +0100, Nick Piggin wrote:
> > > anything when changing the pte to be _more_ permissive, and I don't
> >
> > Note that in my patch
On Wed, Feb 20, 2008 at 02:00:38AM +0100, Andrea Arcangeli wrote:
> On Wed, Feb 20, 2008 at 10:08:49AM +1100, Nick Piggin wrote:
> > You can't sleep inside rcu_read_lock()!
> >
> > I must say that for a patch that is up to v8 or whatever and is
> > posted twice a week to such a big cc list, it is
On Wed, Feb 20, 2008 at 10:55:20AM +1100, Nick Piggin wrote:
> On Friday 15 February 2008 17:49, Christoph Lameter wrote:
> > These special additional callbacks are required because XPmem (and likely
> > other mechanisms) do use their own rmap (multiple processes on a series
> > of remote Linux ins
On Wed, Feb 20, 2008 at 02:11:41PM +1100, Nick Piggin wrote:
> On Wednesday 20 February 2008 14:00, Robin Holt wrote:
> > On Wed, Feb 20, 2008 at 02:00:38AM +0100, Andrea Arcangeli wrote:
> > > On Wed, Feb 20, 2008 at 10:08:49AM +1100, Nick Piggin wrote:
>
> > >
On Wed, Feb 20, 2008 at 02:51:45PM +1100, Nick Piggin wrote:
> On Wednesday 20 February 2008 14:12, Robin Holt wrote:
> > For XPMEM, we do not currently allow file backed
> > mapping pages from being exported so we should never reach this condition.
> > It has been an issue
On Wed, Feb 20, 2008 at 03:00:36AM -0600, Robin Holt wrote:
> On Wed, Feb 20, 2008 at 02:51:45PM +1100, Nick Piggin wrote:
> > On Wednesday 20 February 2008 14:12, Robin Holt wrote:
> > > For XPMEM, we do not currently allow file backed
> > > mapping pages from being
On Wed, Feb 20, 2008 at 11:39:42AM +0100, Andrea Arcangeli wrote:
> Given Nick's comments I ported my version of the mmu notifiers to
> latest mainline. There are no known bugs AFIK and it's obviously safe
> (nothing is allowed to schedule inside rcu_read_lock taken by
> mmu_notifier() with my patc
On Wed, Feb 20, 2008 at 01:03:24PM +0100, Andrea Arcangeli wrote:
> I'm unconvinced both the main linux VM and the mmu notifier should be
> changed like this just to support xpmem. All non-sleeping users don't
> need that. Nevertheless I'm fully welcome to support xpmem (and it's
> not my call nor
On Wed, Feb 20, 2008 at 01:32:36PM +0100, Andrea Arcangeli wrote:
> On Wed, Feb 20, 2008 at 06:24:24AM -0600, Robin Holt wrote:
> > We do not need to do any allocation in the messaging layer, all
> > structures used for messaging are allocated at module load time.
> > The allo
On Wed, Feb 20, 2008 at 11:39:42AM +0100, Andrea Arcangeli wrote:
> XPMEM simply can't use RCU for the registration locking if it wants to
> schedule inside the mmu notifier calls. So I guess it's better to add
Whoa there. In Christoph's patch, we did not use rcu for the list. It
was a simple hl
On Thu, Feb 21, 2008 at 03:20:02PM +1100, Nick Piggin wrote:
> > > So why can't you export a device from your xpmem driver, which
> > > can be mmap()ed to give out "anonymous" memory pages to be used
> > > for these communication buffers?
> >
> > Because we need to have heap and stack available as
On Tue, Feb 26, 2008 at 07:52:41PM +1100, Nick Piggin wrote:
> On Tuesday 26 February 2008 18:21, Gleb Natapov wrote:
> > On Tue, Feb 26, 2008 at 05:11:32PM +1100, Nick Piggin wrote:
> > > > You are missing one point here. The MPI specifications that have
> > > > been out there for decades do not
> > That is it. That is all our allowed interaction with the users process.
>
> OK, when you said something along the lines of "the MPT library has
> control of the comm buffer", then I assumed it was an area of virtual
> memory which is set up as part of initialization, rather than during
> runt
On Thu, Feb 28, 2008 at 01:52:50AM +0100, Andrea Arcangeli wrote:
> On Wed, Feb 27, 2008 at 04:14:08PM -0800, Christoph Lameter wrote:
> > Erm. This would also be needed by RDMA etc.
>
> The only RDMA I know is Quadrics, and Quadrics apparently doesn't need
> to schedule inside the invalidate meth
On Wed, Mar 05, 2008 at 07:09:55AM +0200, Avi Kivity wrote:
> Isn't that out of the question for .25?
I keep hearing this mantra. What is so compelling about the .25
release? When seems to be more important than what. While I understand
product release cycles, etc. and can certainly agree with
On Wed, Apr 02, 2008 at 08:49:52AM +0200, Andrea Arcangeli wrote:
> Most other patches will apply cleanly on top of my coming mmu
> notifiers #v10 that I hope will go in -mm.
>
> For #v10 the only two left open issues to discuss are:
Does your v10 allow sleeping inside the callbacks?
Thanks,
Rob
1 - 100 of 109 matches
Mail list logo